In-Home Technology for Caregivers of People With Dementia and Mild Cognitive Impairment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for In-Home Technology for Caregivers of People With Dementia and Mild Cognitive Impairment (NCT04206670), a randomized na study. The primary outcome is 3 Months Assessment for Center for Epidemiological Studies Depression Scale (CES-D).

2. Clinical Question

3. Trial Design

The registry describes a randomized, single-masked, parallel-group study. Participants were assigned to the In-Home Technology System or Waiting Control. The record does not report who was masked, so the masking label cannot establish whether participants, outcome assessors, or other study personnel were unaware of group assignment. The registry lists an enrollment of 216 and a completed study status.

The study aimed to develop and evaluate in-home supportive technology for caregivers of people with Alzheimer’s disease, other dementias, or mild cognitive impairment. The stated aims concern anxiety, burden, and loneliness, while the posted primary outcome analyzed is CES-D score at three months after baseline. The record lists eight outcome measures, but it does not provide posted results for secondary outcomes.

The intervention group was to receive the full system and remote installation assistance. The system description names entry, activity, temperature, and water-leak sensors, voice control, a digital display, and a local router. The waiting group was assigned a date for system installation six months after entering the study and was to complete questionnaires during the waiting period. The registry record does not report the rationale for this timing or further details of the intervention schedule.

The study began on 2020-03-10, reached primary completion on 2021-12-08, and is marked completed. The registry record does not report the original design assumptions, the planned analysis population, or whether any interim analyses were conducted. The participant counts in the primary outcome results are 125 for the technology group and 61 for the waiting group. These are the counts reported for this outcome, not a description of every participant’s status throughout the trial.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Note that outcomes are only assessed regarding caregiver health and well-being. Patients did not provide any information; therefore, all information reported for demographics and measures is only regarding caregivers.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as 3 Months Assessment for Center for Epidemiological Studies Depression Scale (CES-D), measured over 3 months after baseline.

The primary outcome is the mean score on the Center for Epidemiological Studies Depression Scale (CES-D) at three months after baseline. The questionnaire is described as having 20 items, each rated on a 0-3 scale; the item scores are summed to produce a total from 0-60. There are no subscales. Higher scores represent worse outcomes, and the record notes that a clinical cut-off is usually set at a score of 16.

A continuous score retains information about the degree of reported symptoms rather than reducing every participant to a category. Comparing group means gives a concise estimate of the average score difference between assignments. It does not show how individual scores were distributed or whether every participant experienced a change in the same direction. The standard deviations help describe variation in scores within each group.

The registry identifies CES-D as a measure of depression, and the study’s stated aims include addressing caregiver anxiety, burden, and loneliness. The record does not report a separate outcome definition linking this score to those other aims, nor does it give an interpretation of what a particular between-group difference would mean in daily life. The result should therefore be read as a comparison of group-average questionnaire scores at the specified assessment time.

7. Results Posted in the Registry

• In-Home Technology System: mean 16.05 versus 18.39 on control; difference -2.34 (95% CI -5.49 to 0.81), standardized effect d = -0.23; z-test p = 0.145.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean difference describe?

The CES-D is a continuous score, so a group mean describes the average score among participants with a reported result in that group. The posted mean difference subtracts the Waiting Control mean from the In-Home Technology System mean. A negative value therefore indicates a lower average score in the technology group, consistent with the scale’s direction in which higher scores represent worse outcomes. This is a comparison of group averages, not a description of every participant’s score.

In general, a mean difference is easy to express in the original scale units and can be compared with the scale’s range and interpretation. It can, however, conceal variation: two groups can have different means while their individual score distributions overlap considerably. The standard deviation describes dispersion around a group mean. It is not the uncertainty of the mean itself, and it should not be confused with the confidence interval for the between-group difference.

How should the confidence interval be understood?

In general, a confidence interval gives a range of values reasonably compatible with the data under the statistical procedure used. It conveys estimation uncertainty, not the range of individual participants’ outcomes. The posted interval for the mean difference runs from -5.487 to 0.807. Because it crosses zero, the data are compatible with no mean difference as well as with differences in either direction. The breadth of the interval also indicates that the estimate is not precise enough to pin down a narrow range of plausible effects.

The registry record does not report the method used to construct this interval or whether the posted comparison was adjusted for other variables. The comparison information is an unadjusted recalculation from the posted group summary statistics. The discrepancy indicator is false, so the record does not flag a conflict with a sponsor analysis. A sponsor analysis takes precedence when the record flags a discrepancy; that situation is not indicated for this result.

What does the p-value add?

The unadjusted p-value is 0.145. In general, a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be if that null model and its assumptions were true. It is not the probability that the null hypothesis is true, nor is it a measure of the size or importance of a result. The registry record does not report the detailed testing procedure or the complete statistical analysis plan, so interpretation should stay with the posted unadjusted comparison.

The p-value and confidence interval address related but distinct aspects of uncertainty. The interval shows the range of estimated differences compatible with the procedure; the p-value summarizes evidence against a null difference under a test. Neither statistic says that the observed result will necessarily recur in another study. Neither establishes the reason for any observed group difference.

Why does variability matter?

In general, greater variability within groups makes a mean difference harder to estimate precisely, while more observations can improve precision when other design features are comparable. The reported standard deviations are 10.3 in the technology group and 10.27 in the waiting group. The posted comparison also reports a pooled standard deviation of 10.29. Pooling is a way to summarize within-group variability for a standardized comparison; the mean difference itself remains in CES-D units.

Over-dispersion is relevant in analyses of counts when observed variation exceeds what a simple count model expects. This primary endpoint is a continuous score, not a count outcome, and the registry record does not report an over-dispersion analysis. General explanations of count-model over-dispersion therefore do not describe the method used for this posted mean comparison.

What can randomization support?

In general, random assignment aims to make comparison groups similar on average in factors known and unknown at assignment. It supports interpreting a between-group contrast as a comparison of assigned strategies, subject to how participants were followed and analyzed. Randomization does not guarantee that all baseline characteristics are balanced in a particular trial, and the registry record does not report baseline CES-D values or a baseline characteristics summary for this comparison.

The trial is described as single-masked, but the record does not identify who was masked. It also does not report the statistical analysis plan, handling of missing scores, or the population included in the analysis. Those omissions limit assessment of whether the posted summary fully reflects the randomized comparison.

10. Confidence Intervals and Precision

The estimated mean difference was -2.34 CES-D units, with a confidence interval from -5.487 to 0.807. The interval includes zero and covers estimates on both sides of it. Thus, the posted data do not determine the direction of the underlying group difference with high precision. The result is compatible with lower mean scores in the technology group, little or no difference, or a small difference in the other direction.

Precision concerns the range of plausible values for the group contrast, not how consistent individual participants’ scores were. The reported standard deviations, 10.3 for the technology group and 10.27 for the waiting group, describe participant-level spread within groups. They do not mean that the group means themselves have that degree of uncertainty. The confidence interval is the relevant posted summary of uncertainty in the mean difference.

The unadjusted p-value of 0.145 is consistent with the interval’s inclusion of zero, but it does not tell readers which values in the interval are more likely or whether a difference matters in practice. The registry record does not report a threshold for a meaningful CES-D difference. Although it notes a usual clinical cut-off of 16, that is a reference for interpreting scores, not a reported threshold for the between-group mean difference.

The record does not report baseline outcome values, the number of missing scores, or the method used to form the confidence interval. These details would help assess how the analyzed data and analysis choices affect precision. Accordingly, the interval describes uncertainty in the posted comparison but cannot resolve questions about unreported data or analysis decisions.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrative calculations based on the observed effect, rather than the trial’s original design justification. The registry record does not report the trial’s own sample-size assumptions. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect.

The 80% power column gives the illustrative number of participants per arm for a design that would have 80% power under those assumptions and the selected effect. The 90% power column gives the corresponding number for 90% power. Power is the probability of detecting an effect of the assumed size under the planning model if that effect is the true one; it is not a guarantee that a study will produce a statistically significant result.

For planning at 100% of the observed effect, the figures are 304 participants per arm for 80% power and 407 participants per arm for 90% power. Planning for 75% of the observed effect gives 540 and 723 participants per arm, respectively. Planning for 50% of the observed effect gives 1,215 and 1,626 participants per arm, respectively. Smaller effects require more participants under the same assumptions because they are harder to distinguish from random variation.

These calculations are not observed power. They are planning illustrations using the observed estimate as a starting point; the observed estimate itself is uncertain, as reflected in its confidence interval. The record does not report whether these assumptions were used when the trial was designed, and the illustrative figures should not be read as a reconstruction of the original design.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -2.340; pooled standard deviation σ = 10.290.
2. n = 2 × (1.96 + 0.842)² × 10.290² / -2.340² = 303.6, rounded up to 304 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -2.34                      # observed difference in means
sigma <- 10.2902270013411                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(4206670)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 304 (80% power), 407 (90% power)
 75% of observed effect: n per arm = 540 (80% power), 723 (90% power)
 50% of observed effect: n per arm = 1,215 (80% power), 1,626 (90% power)
power.t.test: 304.5 
Simulated power at n = 304 per arm: 0.797 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures, while the number of posted secondary outcomes is zero. The record reports this primary CES-D comparison but does not provide secondary outcome results. It does not report how many hypotheses were prespecified for analysis, how the primary outcome was distinguished from other planned outcomes, or whether any multiplicity adjustment was planned or used.

In general, when multiple outcomes or comparisons are examined, the chance of finding an apparently unusual result by chance can increase. Methods for controlling that issue depend on the planned family of comparisons and the analysis strategy. Because the registry record does not report those details, it is not possible to determine whether multiplicity was addressed or to describe a specific correction. The unadjusted p-value should be interpreted as the result for the posted comparison, not as evidence that a broader set of outcomes was tested in a particular way.

The record does not report the number of missing CES-D assessments, reasons for missingness, or how missing scores were handled. It also does not report whether the analysis included all randomized participants, only participants with observed follow-up scores, or another analysis population. The group counts attached to the outcome are 125 and 61, but the registry does not explain how those counts relate to missing data or the randomized population.

In general, missing outcome data can affect both the precision and interpretation of a comparison, particularly if the likelihood of missingness is related to outcome values or group assignment. Without the amount, pattern, and handling of missing data, the direction or size of any resulting bias cannot be assessed from this record. The analysis population and missing-data approach are not reported in the registry record.

The record also does not report a detailed statistical analysis plan, adjusted analyses, or interim analyses. These omissions do not establish that such procedures were absent; they mean that they are not documented in the registry record. Interpretation therefore remains limited to the posted unadjusted outcome summaries and comparison.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports serious adverse events for 148 participants at risk in the In-Home Technology System group and 68 at risk in the Waiting Control group. The reported number affected by serious adverse events is 0 in each group. The record also reports 0 participants affected by other adverse events in each group, and 0 deaths in each group.

These are the safety figures recorded for the listed groups. They do not mean that every possible event was ruled out, and they do not describe event types beyond the categories and counts reported. The registry lists no serious-event terms. It does not report how events were collected, assessed, or followed over time, so the figures should be read as the posted registry counts rather than as a complete account of all safety monitoring.

The participants at risk reported for safety differ from the counts attached to the CES-D outcome. The registry record does not explain the difference, so no reason for it can be inferred. These safety counts also do not indicate the number of people with completed outcome questionnaires or define the analysis population for the primary result.

15. What the Result Does, and Does Not, Mean

The posted result describes a lower mean CES-D score in the In-Home Technology System group than in the Waiting Control group at three months after baseline. The estimated unadjusted mean difference was -2.34 units. The confidence interval includes zero, and the unadjusted p-value was 0.145, so the result leaves substantial uncertainty about the size and direction of the underlying group difference.

This is not proof that the groups had the same mean, nor does it establish a definite difference. A range of values remains compatible with the data under the confidence-interval procedure. The record does not report a threshold for a meaningful change in the group mean, so statistical uncertainty cannot be translated into a judgment about practical importance using a trial-specific criterion.

The measure is a questionnaire score, with higher scores representing worse outcomes. The group means do not indicate how many caregivers had scores above the usual clinical cut-off of 16, because individual score distributions and threshold counts are not reported. Nor does the mean comparison show whether individual participants improved or worsened from baseline; baseline scores and participant-level changes are not reported.

The result applies to this primary outcome and this assessment time. It does not establish a result for caregiver anxiety, burden, loneliness, or other outcomes named in the study aims. Secondary results are not posted in the registry record. The comparison also does not provide evidence about later outcomes or long-term experience with the system.

Finally, the registry result is an unadjusted comparison. The record does not report missing-data handling, the analysis population, or baseline outcome values. Those limits matter when interpreting how fully the posted estimate represents the randomized comparison. The appropriate conclusion is a description of the observed score difference and its uncertainty, rather than a claim about treatment benefit or absence of benefit.

16. Why This Trial Matters Statistically

This trial is statistically instructive because the outcome is a continuous questionnaire score and the principal comparison is expressed in the score’s original units. The mean difference of -2.34 is directly interpretable as a difference in average CES-D scores, while the confidence interval shows that the estimate is uncertain and includes zero. Considering the estimate together with its interval is more informative than treating the p-value as a yes-or-no verdict.

The result also illustrates why a group mean should not be confused with individual experience. The CES-D scores have meaningful variation within both groups, as reflected in the reported standard deviations. A difference between averages does not indicate that every caregiver had a score change in that direction, and the group mean cannot identify how many individuals were above the stated clinical cut-off.

The posted planning illustrations make the role of effect size visible: under their stated assumptions, planning for a smaller fraction of the observed effect requires more participants per arm. This is a general feature of statistical planning, not evidence that the trial was designed using those calculations. The record does not report its original sample-size justification.

Finally, the registry record highlights the importance of transparent reporting. It gives group summaries and an unadjusted comparison, but does not report the analysis population, missing-data handling, baseline CES-D scores, or details of multiplicity and the analysis plan. Those details help readers evaluate how an estimate was produced and what questions it can answer. Their absence in the registry record is a limit on interpretation, not proof that no related procedures occurred.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report baseline CES-D scores, so the primary result cannot show whether groups began with similar average scores or how scores changed over time. It gives a follow-up comparison at three months after baseline but does not report participant-level change or a baseline-adjusted estimate. The record also does not describe the confidence-interval method or a detailed analysis model.

Missing outcome data are another unresolved issue. The record does not report how many CES-D assessments were missing, why they were missing, how missing values were handled, or which analysis population was used. The counts reported for the primary outcome differ from the safety at-risk counts, but the registry record does not explain why. No explanation should be inferred from those figures alone.

The study is labeled single-masked, but the record does not identify who was masked. It does not report whether outcome assessment was blinded or how the self-reported questionnaire was administered. The record also does not report the original sample-size assumptions, multiplicity procedures, or whether interim analyses occurred. These details limit evaluation of the design and analysis choices.

The outcome result is limited to CES-D at three months after baseline. Although the study aims mention anxiety, burden, and loneliness and the record lists eight outcome measures, no secondary results are posted. The record does not report longer-term results for the primary comparison. It also does not provide individual score distributions, so the clinical cut-off cannot be used to infer how many participants fell above or below it.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: In-Home Technology for Caregivers of People With Dementia and Mild Cognitive Impairment — NCT04206670.
• Registry record last updated 2022-05-10; status last verified 2022-04.
• Results first posted 2022-05-10.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

The registry record describes a completed randomized, single-masked, parallel-group study of in-home supportive technology for caregivers of people with dementia, Alzheimer’s disease, or mild cognitive impairment. Enrollment is listed as 216. The primary outcome is mean CES-D score three months after baseline, with higher scores representing worse outcomes.

The technology group had a reported mean score of 16.05 and standard deviation of 10.3; the Waiting Control group had a mean of 18.39 and standard deviation of 10.27. The unadjusted mean difference was -2.34, with a confidence interval from -5.487 to 0.807 and an unadjusted p-value of 0.145. No secondary outcome results are posted. The registry record does not report the analysis population, missing-data handling, or original sample-size assumptions.